In [25]:
import pandas as pd
data = pd.read_csv("/Users/maruthiganesh/Desktop/B.TECH /My Projects/MUTUAL FUND ANALYSIS/nifty50_closing_prices.csv")
print(data.head())
Date RELIANCE.NS HDFCBANK.NS ICICIBANK.NS \
0 2024-08-20 00:00:00+05:30 2991.899902 1637.699951 1179.449951
1 2024-08-21 00:00:00+05:30 2997.350098 1625.800049 1174.849976
2 2024-08-22 00:00:00+05:30 2996.250000 1631.300049 1191.099976
3 2024-08-23 00:00:00+05:30 2999.949951 1625.050049 1203.500000
4 2024-08-26 00:00:00+05:30 3025.199951 1639.949951 1213.300049
INFY.NS TCS.NS KOTAKBANK.NS HINDUNILVR.NS ITC.NS \
0 1872.199951 4523.299805 1805.650024 2751.050049 498.799988
1 1872.699951 4551.500000 1812.949951 2791.199951 505.399994
2 1880.250000 4502.000000 1821.500000 2792.800049 504.549988
3 1862.099976 4463.899902 1818.000000 2815.600098 505.799988
4 1876.150024 4502.450195 1812.500000 2821.149902 505.700012
LT.NS ... HEROMOTOCO.NS DRREDDY.NS SHREECEM.NS BRITANNIA.NS \
0 3572.699951 ... 5244.399902 6965.350098 24730.550781 5765.799805
1 3596.050049 ... 5284.700195 7062.450195 24808.050781 5837.350098
2 3606.500000 ... 5329.950195 6969.049805 25012.400391 5836.799805
3 3598.550049 ... 5384.899902 6954.500000 24706.050781 5792.649902
4 3641.899902 ... 5343.750000 6943.299805 24906.449219 5796.950195
UPL.NS EICHERMOT.NS SBILIFE.NS ADANIPORTS.NS BAJAJ-AUTO.NS \
0 566.150024 4883.250000 1761.300049 1492.550049 9779.700195
1 568.299988 4913.549805 1800.599976 1503.500000 9852.000000
2 579.150024 4933.549805 1795.250000 1492.300049 9914.200195
3 573.700012 4898.100098 1789.300049 1491.300049 10406.450195
4 577.450012 4875.200195 1796.250000 1482.550049 10432.549805
HINDALCO.NS
0 672.900024
1 685.599976
2 685.549988
3 685.099976
4 711.849976
[5 rows x 51 columns]
In [3]:
data['Date'] = pd.to_datetime(data['Date'])
In [35]:
print(data.isnull().sum())
num_companies = len(data.columns) - 1
print("Number of Companies = " + str(num_companies))
Date 0 RELIANCE.NS 0 HDFCBANK.NS 0 ICICIBANK.NS 0 INFY.NS 0 TCS.NS 0 KOTAKBANK.NS 0 HINDUNILVR.NS 0 ITC.NS 0 LT.NS 0 SBIN.NS 0 BAJFINANCE.NS 0 BHARTIARTL.NS 0 HCLTECH.NS 0 ASIANPAINT.NS 0 AXISBANK.NS 0 DMART.NS 0 MARUTI.NS 0 ULTRACEMCO.NS 0 HDFC.NS 24 TITAN.NS 0 SUNPHARMA.NS 0 M&M.NS 0 NESTLEIND.NS 0 WIPRO.NS 0 ADANIGREEN.NS 0 TATASTEEL.NS 0 JSWSTEEL.NS 0 POWERGRID.NS 0 ONGC.NS 0 NTPC.NS 0 COALINDIA.NS 0 BPCL.NS 0 IOC.NS 0 TECHM.NS 0 INDUSINDBK.NS 0 DIVISLAB.NS 0 GRASIM.NS 0 CIPLA.NS 0 BAJAJFINSV.NS 0 TATAMOTORS.NS 0 HEROMOTOCO.NS 0 DRREDDY.NS 0 SHREECEM.NS 0 BRITANNIA.NS 0 UPL.NS 0 EICHERMOT.NS 0 SBILIFE.NS 0 ADANIPORTS.NS 0 BAJAJ-AUTO.NS 0 HINDALCO.NS 0 dtype: int64 Number of Companies = 50
In [5]:
data.ffill(inplace=True)
In [6]:
import plotly.graph_objs as go
import plotly.express as px
fig = go.Figure()
for company in data.columns[1:]:
fig.add_trace(go.Scatter(x=data['Date'], y=data[company],
mode='lines',
name=company,
opacity=0.5))
fig.update_layout(
title='Stock Price Trends of All Indian Companies',
xaxis_title='Date',
yaxis_title='Closing Price (INR)',
xaxis=dict(tickangle=45),
legend=dict(
x=1.05,
y=1,
traceorder="normal",
font=dict(size=10),
orientation="v"
),
margin=dict(l=0, r=0, t=30, b=0),
hovermode='x',
template='plotly_white'
)
In [7]:
all_companies = data.columns[1:]
volatility_all_companies = data[all_companies].std() #.std() calculates the standard deviation of the data
volatility_all_companies.sort_values(ascending=False).head(10) #
Out[7]:
BAJAJ-AUTO.NS 659.810841 SHREECEM.NS 429.919834 BAJFINANCE.NS 306.658594 DIVISLAB.NS 247.674895 HEROMOTOCO.NS 247.092728 DRREDDY.NS 175.124908 ULTRACEMCO.NS 172.673053 DMART.NS 155.593701 BRITANNIA.NS 144.164343 MARUTI.NS 109.587342 dtype: float64
In [9]:
growth_all_companies = data[all_companies].pct_change(fill_method=None) * 100 #.pct_change() in pandas to calculate the percentage change between the current and a prior element.
average_growth_all_companies = growth_all_companies.mean() #.mean() is used to find the mean of the data
average_growth_all_companies.sort_values(ascending=False).head(10)
Out[9]:
BAJAJ-AUTO.NS 0.883421 BAJAJFINSV.NS 0.791730 BHARTIARTL.NS 0.735219 DIVISLAB.NS 0.634851 HEROMOTOCO.NS 0.602192 ICICIBANK.NS 0.557742 BAJFINANCE.NS 0.536819 TITAN.NS 0.393800 HINDUNILVR.NS 0.351634 BRITANNIA.NS 0.327747 dtype: float64
----- Calculating the Percentage Change in the the Initial and Final Values to calculate the Return on Inveestment -----
In [10]:
initial_prices_all = data[all_companies].iloc[0]
final_prices_all = data[all_companies].iloc[-1]
roi_all_companies = ((final_prices_all - initial_prices_all) / initial_prices_all) * 100
roi_all_companies.sort_values(ascending=False).head(10)
Out[10]:
BAJAJ-AUTO.NS 22.107017 BAJAJFINSV.NS 19.642973 BHARTIARTL.NS 18.120965 DIVISLAB.NS 15.404976 HEROMOTOCO.NS 14.660402 ICICIBANK.NS 13.480860 BAJFINANCE.NS 12.797149 TITAN.NS 9.275089 HINDUNILVR.NS 8.235039 BRITANNIA.NS 7.713587 dtype: float64
----- Creating a Mutual Fund Plan Based on High ROI and Low Risk ----- --> We will set thresholds for ROI and volatility to select companies that provide good returns with lower risks. --> Rank all companies based on their ROI and volatility scores. --> Allocate more investment to companies with higher ROI and lower volatility.
In [18]:
roi_threshold = roi_all_companies.median()
volatility_threshold = volatility_all_companies.median()
selected_companies = roi_all_companies[(roi_all_companies > roi_threshold) & (volatility_all_companies < volatility_threshold)]
selected_companies.sort_values(ascending=False)
Out[18]:
ICICIBANK.NS 13.480860 INDUSINDBK.NS 7.159914 JSWSTEEL.NS 7.021748 AXISBANK.NS 6.592466 HDFCBANK.NS 6.319839 SUNPHARMA.NS 5.627425 KOTAKBANK.NS 5.474481 CIPLA.NS 4.850117 NTPC.NS 4.356926 dtype: float64
----- To balance the investment between these companies, we can use an inverse volatility ratio for allocation. Companies with lower volatility will get a higher weight -----
In [19]:
selected_volatility = volatility_all_companies[selected_companies.index]
inverse_volatility = 1 / selected_volatility
investment_ratios = ( inverse_volatility / inverse_volatility.sum() )*100
investment_ratios.sort_values(ascending=False)
Out[19]:
NTPC.NS 28.076767 JSWSTEEL.NS 15.998503 AXISBANK.NS 9.223133 HDFCBANK.NS 8.933035 CIPLA.NS 8.478347 KOTAKBANK.NS 7.664235 INDUSINDBK.NS 7.443153 SUNPHARMA.NS 7.255261 ICICIBANK.NS 6.927566 dtype: float64
In [20]:
top_growth_companies = average_growth_all_companies.sort_values(ascending=False).head(10)
risk_growth_rate_companies = volatility_all_companies[top_growth_companies.index]
risk_mutual_fund_companies = volatility_all_companies[selected_companies.index]
fig = go.Figure()
fig.add_trace(go.Bar(
y=risk_mutual_fund_companies.index,
x=risk_mutual_fund_companies,
orientation='h', # Horizontal bar
name='Mutual Fund Companies',
marker=dict(color='blue')
))
fig.add_trace(go.Bar(
y=risk_growth_rate_companies.index,
x=risk_growth_rate_companies,
orientation='h',
name='Growth Rate Companies',
marker=dict(color='green'),
opacity=0.7
))
fig.update_layout(
title='Risk Comparison: Mutual Fund vs Growth Rate Companies',
xaxis_title='Volatility (Standard Deviation)',
yaxis_title='Companies',
barmode='overlay',
legend=dict(title='Company Type'),
template='plotly_white'
)
fig.show()
In [22]:
expected_roi_mutual_fund = roi_all_companies[selected_companies.index]
expected_roi_growth_companies = roi_all_companies[top_growth_companies.index]
fig = go.Figure()
fig.add_trace(go.Bar(
y=expected_roi_mutual_fund.index,
x=expected_roi_mutual_fund,
orientation='h',
name='Mutual Fund Companies',
marker=dict(color='blue')
))
fig.add_trace(go.Bar(
y=expected_roi_growth_companies.index,
x=expected_roi_growth_companies,
orientation='h',
name='Growth Rate Companies',
marker=dict(color='green'),
opacity=0.7
))
fig.update_layout(
title='Expected ROI Comparison: Mutual Fund vs Growth Rate Companies',
xaxis_title='Expected ROI (%)',
yaxis_title='Companies',
barmode='overlay',
legend=dict(title='Company Type'),
template='plotly_white'
)
fig.show()
----- Calculating Expected Returns ----- Calculating the expected returns a person will get from the mutual fund if he/she invests ₹5000 every month. To calculate the expected value a person will accumulate over 1 year, 3 years, 5 years, and 10 years through the mutual fund plan. ---> We sssume the person is investing 5000 rupees every month. ---> We use the expected ROI from the mutual fund companies to simulate the growth over time. ---> Compute the compounded value of the investments for each period (1y, 3y, 5y, and 10y). ---> Visualize the accumulated value over these periods.
In [23]:
import numpy as np
monthly_investment = 5000 # Monthly investment in INR
years = [1, 3, 5, 10] # Investment periods (in years)
n = 12 # Number of times interest is compounded per year (monthly)
avg_roi = expected_roi_mutual_fund.mean() / 100 # Convert to decimal
def future_value(P, r, n, t):
return P * (((1 + r/n)**(n*t) - 1) / (r/n)) * (1 + r/n)
future_values = [future_value(monthly_investment, avg_roi, n, t) for t in years]
fig = go.Figure()
fig.add_trace(go.Scatter(
x=[str(year) + " year" for year in years],
y=future_values,
mode='lines+markers',
line=dict(color='blue'),
marker=dict(size=8),
name='Future Value'
))
fig.update_layout(
title="Expected Value of Investments of ₹ 5000 Per Month (Mutual Funds)",
xaxis_title="Investment Period",
yaxis_title="Future Value (INR)",
xaxis=dict(showgrid=True, gridcolor='lightgrey'),
yaxis=dict(showgrid=True, gridcolor='lightgrey'),
template="plotly_white",
hovermode='x'
)
fig.show()